Production and productivity can be boosted either through increased use of inputs and/or improvement in technology or by improving the efficiency of producers or firms, given fixed level of inputs and technology. Even though agriculture stays the main stay of Ethiopian economy, level of agricultural productivity in general and crop productivity in particular is very low. Out of the total grain production in Ethiopia, cereals account for roughly 60 percent of rural employment and 80 percent of total cultivated land. However, Yield of cereals has been consistently well below world and even of least developing countries average yield, indicating poor productivity of the crops in the country. Given capital constraint in the country, it is difficult to adopt new technology to enhance productivity. Hence, working to improve production efficiency is best option on hand. As a result, there are a number of studies done on area of efficiency analysis in Ethiopia. However, the novelty of this study can be explained by three facts. First of all it has used national data, collected by International Food Policy Research Institute (IFPRI), with enough number of observations to do plot level analysis considering biological factors that determine inefficiency. Second, efficiency analysis is not based on a single crop rather on major crops in general as well as teff, wheat and maize independently. Last but not least, the study employed one stage approach in which both technical efficiency and factors of inefficiency are analyzed simultaneously. Therefore, this study was done to evaluate the efficiency and identify factors that explain the variation in inefficiency of crop production in Ethiopia. This study principally used the 2009 Ethiopia Rural Household Survey (ERHS) which is collected by IFPRI. As far as analysis is concerned, both descriptive and econometric methods were used. Descriptive statistics (mean, percentage, range, etc.) is used to summarize the variables in the model and describe the study area. Econometric model, Stochastic Production Frontier model, is used to estimate the elasticity of production function, determine the determinants of inefficiency and estimate the level of efficiency. Given that we are considering a developing country setting where by the main concern is output shortfall rather than input over use, preference has been given to primal or output oriented approach of measuring efficiency. In this study, effort was made to test the hypotheses before rushing to interpret the model outputs. First, the γ parameter estimates of all production functions were significant at 5% significance level, indicating Stochastic Frontier Production function is more appropriate than convectional production function or there is significant technical inefficiency variation among plots. The γ value of 0.636 for the major crops production function can be then interpreted as, 63% of the variation in output among plots is explained by technical inefficiency. Similarly, variation in out put due to technical inefficiency for teff, wheat and maize production were calculated to be 88.5, 45.5 and 77.8 percent respectively. The second step, following the existence of inefficiency, is to check if there exist one or more variables that could explain the variation in technical inefficiency. Log likelihood ratio was used to test the hypothesis. Accordingly, all calculated LL ratio values were greater than the critical value of LL ratio, with upper 5 % level of significance. Hence, the null hypotheses that determinant variables in the inefficiency effect model are simultaneously equal to zero are rejected. In other words, there exists at least one explanatory variable that explains the variation in the technical inefficiency among plots. The ML estimate results shown that, all variables were found to be binding in the production of major crops, meaning that an increase in one of inputs will enhance output keeping everything constant. As far as teff production is concerned, only land was a significant variable that explains the variation in teff output among plots. Land, DAP and seed were found to have significant and positive effect in wheat production. According to result of this study, land and seed were major determinants of maize production in Ethiopia. Generally, all significant input variables were found to affect output positively, as expected. Moreover, the model output depicted that the mean level of TE for major crops, Teff, Wheat and Maize production was found to be 63.56, 67.26, 84.16 and 91.41 percent, respectively. The inefficiency effect analysis shown that, age of the household head measured in years was found to be the determinant of technical inefficiency, of teff production and education was found to have negative and significant effect on major crops and wheat technical inefficiency (1% significance level). Knowledge about land policy was found to have significant and negative effect on technical inefficiency of wheat production (1% significance level). Similarly, participation in soil and water conservation activities was found to have negative and significant effect on technical inefficiency of major crops and wheat production. In this study frequency of extension contact was found to have unexpected and strange result; the more frequently the farmers meet extension workers the more it competes their time to do agricultural activities. The result of this study also confirmed as rich farmers are relatively less inefficient than poor once, in major crops production, and fertile plots of wheat are significantly less inefficient than infertile once. Similarly, flat teff and maize plots are more efficient than otherwise. The other plot specific variable that was found to have negative and significant effect on technical inefficiency of major crop production was adoption of improved seed. The last but not least, variable that explains variation in inefficiency was found to be livestock ownership. Generally, results of this study confirmed that there is a room to enhance productivity by improving the efficiency of production, given same level of input and current technology.